Green Global
Foundation   Journal
...an emerging window for scientific research and publication across the globe

STUDY ON CLASSIFICATION OF TONGUE MOVEMENT EAR PRESSURE (TMEP) SIGNALS FOR HUMAN MACHINE INTERFACE (HMI) (Pages : 75 - 78)

Few projects on human machine interface (HMI) are already running successfully as like, power while chair, prosthetics etc. and proved to very resourceful invention for mankind with a great field for improvement. The goal of this research project is to describe pattern classification of tongue movement action by Tongue Movement Ear Pressure (TMEP) signals. The classification has been done on four tongue movement actions (up, down, left, and right). Cross validation method is used for Classifier training and testing to get a better accuracy and Bayesian classifier have been used to get better performance in real-time for hands free communications. The average classification accuracy with the real time interfering signals achieved 82.9% (Bayesian) and its shown that uni-variance Bayesian classifier performs better than multivariate classifier for this new six features mean, median, variance, kurtosis, skewness and entropy. This approach of combining the Bayesian classifier and the k-fold cross validation method provides a robust and efficient method for a real-time assistive human machine interface based on tongue-movement ear pressure signals.Download